Of course, this is a challenge to the philosophy which says that a detailed enough simulation of a brain will create a conscious person, regardless of the computational substrate.
So if we do simulate a brain and it tells you it’s conscious and experiences green (through a camera), would you then agree that there’s no need for dualism?
I don’t think there’s a need for dualism anyway; there’s a need for a new physical ontology. But a simulation of a conscious brain should tell you that it’s conscious even if it’s not, or else it’s not an effective simulation.
So p-zombies are possible, and in humans, the physical processes (of the brain) are somehow “magically” correlated / isomorphic to mental phenomena, whereas this doesn’t happen in simulations, for what (unknown?) reasons?
The p-zombie theory holds that being able to conceive of something makes it possible; and because p-zombies are possible, therefore dualism. The tricky bit appears to be “conceive of” in a sense that implies possibility. Consider these statements:
I can conceive of 2+2=4 being true in conventional Peano arithmetic.
I can conceive of 2+2=5 being true in conventional Peano arithmetic.
I can conceive of P being equal to NP.
I can conceive of P not being equal to NP.
I can conceive of p-zombies, therefore dualism.
If I can conceive of p-zombies then dualism, which is a confused idea, therefore p-zombies is a confused idea by reductio ad absurdum.
With the second, I am claiming to “conceive of” something trivially false. I arguably haven’t conceived of anything actually possible; I’ve just shuffled some words together.
With the third and fourth, I’m claiming to have conceived of something no-one knows (though many suspect 3 is false and 4 is true). To what extent have I actually thought it through? At some point I will hit a contradiction with one of them, though no-one has yet. Both are “conceivable” in some sense; certainly that they’ve formed a sentence in their head that they can try out for its logical implications. But one of those statements is as wrong as 2+2=5 nevertheless.
When someone claims that p-zombies are a conceivable thing at all, and that they have conceived of them, this doesn’t actually say anything about the world or what is even possible; it just says they’ve formed a sentence in their head they think they can try out for its logical implications. But try telling them this. (I have, and haven’t managed a sufficiently robust form of 6. to be convincing.)
(I still consider the fundamental argument in favour of dualism is that its advocates really want it to be true, and that p-zombies is like creationism for smart people.)
I’ve just realised that the second zombie post in the sequences makes exactly the point I made above: the gap between “I don’t see a contradiction yet” and “this is logically possible” and what happens when you conflate the two (for instance, you might think p-zombies aren’t utterly stupid).
The fact that you’re saying that two things can be isomorphic or something close to it, and both say they’re conscious, even though one is and one isn’t — alternately, the claim that you can know things about your substrate just through introspection (why do you think you’re conscious if you think something closely analogous would say so and be wrong?) — seems analogous to saying that zombies can exist, though maybe not as clearly problematic. Does this make you worry?
Running a simulation that doesn’t model already-existing phenomena is like minting counterfeit currency. Knowingly going along with a popular delusion is like knowingly passing along fake bills. But if instead of naive classical simulation, one works with the entangled details of the quantum information theoretic world, then that counts as adding value to the economy and not as counterfeiting. One way of exerting optimization pressure in this way is to selectively collapse parts of the universal wave function. A superintelligence could even do this to what humans call the “past”. That’s kind of a big deal.
-- shit I say on Facebook
I mostly care about this kinda stuff ’cuz I’m afraid of demons (knowingly passing along fake bills).
ETA: I’d like to hear an explanation for the downvotes, for my amusement.
So if we do simulate a brain and it tells you it’s conscious and experiences green (through a camera), would you then agree that there’s no need for dualism?
I don’t think there’s a need for dualism anyway; there’s a need for a new physical ontology. But a simulation of a conscious brain should tell you that it’s conscious even if it’s not, or else it’s not an effective simulation.
So p-zombies are possible, and in humans, the physical processes (of the brain) are somehow “magically” correlated / isomorphic to mental phenomena, whereas this doesn’t happen in simulations, for what (unknown?) reasons?
The p-zombie theory holds that being able to conceive of something makes it possible; and because p-zombies are possible, therefore dualism. The tricky bit appears to be “conceive of” in a sense that implies possibility. Consider these statements:
I can conceive of 2+2=4 being true in conventional Peano arithmetic.
I can conceive of 2+2=5 being true in conventional Peano arithmetic.
I can conceive of P being equal to NP.
I can conceive of P not being equal to NP.
I can conceive of p-zombies, therefore dualism.
If I can conceive of p-zombies then dualism, which is a confused idea, therefore p-zombies is a confused idea by reductio ad absurdum.
With the second, I am claiming to “conceive of” something trivially false. I arguably haven’t conceived of anything actually possible; I’ve just shuffled some words together.
With the third and fourth, I’m claiming to have conceived of something no-one knows (though many suspect 3 is false and 4 is true). To what extent have I actually thought it through? At some point I will hit a contradiction with one of them, though no-one has yet. Both are “conceivable” in some sense; certainly that they’ve formed a sentence in their head that they can try out for its logical implications. But one of those statements is as wrong as 2+2=5 nevertheless.
When someone claims that p-zombies are a conceivable thing at all, and that they have conceived of them, this doesn’t actually say anything about the world or what is even possible; it just says they’ve formed a sentence in their head they think they can try out for its logical implications. But try telling them this. (I have, and haven’t managed a sufficiently robust form of 6. to be convincing.)
(I still consider the fundamental argument in favour of dualism is that its advocates really want it to be true, and that p-zombies is like creationism for smart people.)
I’ve just realised that the second zombie post in the sequences makes exactly the point I made above: the gap between “I don’t see a contradiction yet” and “this is logically possible” and what happens when you conflate the two (for instance, you might think p-zombies aren’t utterly stupid).
Lots of things don’t happen in simulations. Simulated planes don’t fly, and simulted explosions don’t destroy anything..
The fact that you’re saying that two things can be isomorphic or something close to it, and both say they’re conscious, even though one is and one isn’t — alternately, the claim that you can know things about your substrate just through introspection (why do you think you’re conscious if you think something closely analogous would say so and be wrong?) — seems analogous to saying that zombies can exist, though maybe not as clearly problematic. Does this make you worry?
-- shit I say on Facebook
I mostly care about this kinda stuff ’cuz I’m afraid of demons (knowingly passing along fake bills).
ETA: I’d like to hear an explanation for the downvotes, for my amusement.
I didn’t downvote it, but if I’d seen it at zero I might well have done, ’cos it makes no damn sense to me. Please unpack.